The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, such as user-provided API specifications, route definitions, and controller code. This creates a surface for indirect prompt injection if the input contains malicious instructions designed to influence the agent's behavior during documentation generation.
Ingestion points: Processes external code, OpenAPI YAML/JSON files, and route definitions provided by the user in the prompt context.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for handling the untrusted input.
Capability inventory: The skill is limited to text generation and formatting (Markdown, HTML, OpenAPI YAML, Postman JSON). It does not have access to sensitive file-system operations, network requests, or subprocess execution.
Sanitization: No explicit sanitization or validation of the input content is specified.
[SAFE]: The skill performs standard documentation tasks. The curl examples provided in the output structure are intended as static documentation templates for the user and do not involve the execution of commands by the agent.

api-documentation